Towards biopsy-less skin diagnosis

Diagnosing skin disease with topically applied mRNA-detecting fluorescent nanoparticles.

Like Comment

The paper in Nature Biomedical Engineering is here:

At the global level, skin conditions are a leading cause of nonfatal disease burden1, and the management of skin disorders continues to be a significant unmet need. Abnormal scar formation, as occurs with hypertrophic and keloidal scars, is an example of skin disease that causes considerable burden2, because of associated itch, pain, reduced limb mobility, and emotional distress. These abnormal scars are associated with >30% of elective surgeries and 90% of burns3. Diagnosis of these scars is only made after visual recognition, requiring disease to be fully developed before treatment can be considered.

In many other skin disorders, accurate diagnosis requires tissue biopsy, with subsequent processing and staining of the tissue to identify morphologic changes as well as disease biomarkers. As with hypertrophic and keloidal scars, these alterations are generally discovered only after skin disease features become visible, preventing early diagnosis and the possibility for intervention before manifestations are advanced. In addition, biopsy is inconvenient (generally requiring suture removal) and resource-intensive, risks infection, and leaves a scar, which is particularly problematic in individuals prone to hypertrophic scarring and/or keloid formation.

Diagnostic tools for skin diseases that avoid invasive biopsy sampling and allow earlier diagnosis when clinical changes are not yet visible, such as through assessment of gene expression, would be highly desirable.

Nanyang Technological University – Northwestern University collaboration. Through a collaboration program between Northwestern University and Nanyang Technological University, we (Bioengineer Assistant Professor Chenjie Xu) formed a partnership with a chemist, Professor Chad Mirkin, and a dermatologist-scientist, Professor Amy Paller, both from Northwestern University. At the time, we were developing microneedle devices for abnormal scar self-management and thus had expertise in cell, tissue and animal models. We realized that spherical nucleic acid (SNA) technology could be highly versatile, biocompatible, and have sufficient selectivity and sensitivity. Moreover, we envisioned a dual-purpose technology that enabled simultaneous detection (NanoFlares) and silencing of diseased gene biomarkers through antisense/siRNA-SNAs.

NanoFlares have sufficient versatility to detect mRNA targets by creating complementary ‘recognition strands’ that hybridize with its target. The goal was to prove that the NanoFlare tool could detect the presence of disease based on biomarker expression, as well as the response to therapeutic agents non-invasively. The goal was identification of early signs of disease activity at a time of greatest potential response to intervention, through detection of gene expression changes using noninvasive technology.

Use of SNA technology for noninvasive diagnosis and management. SNA nanoconstruct technology confers benefits that include resistance to intracellular nuclease activity, high quantities of cellular uptake, sequence programmability and, importantly, skin penetration properties. The easily programmable DNA sequences facilitate sensor development by identifying DNA sequences to detect target mRNA in a selective and sensitive manner. Strongly anionic NanoFlares and potentially other DNA nanostructures could penetrate tissue barriers and allow diagnosis of other superficial tissue diseases. This approach would facilitate self-applied diagnostic sensors and shift away from resource-intensive biopsy-based diagnostic procedures.

Our personal point-of-view. The most significant impact of transdermal NanoFlare detection is that it has the potential to diagnose disease at the gene expression level, which is potentially early and brings us closer to targeted treatment based on mechanism of disease rather than just clinical diagnosis. Diagnosis using NanoFlares could potentially reduce the cost of diagnosis, while minimizing pain, inconvenience, risk of infection and scarring. The technology even has the potential to be applied by the patient through a kit for subsequent physician assessment. While our work to date evaluated the intensity of the NanoFlares as a surrogate for extent of gene expression, it is possible that the technology could even detect genetic point mutations, not only distinguishing gene expression changes quantitatively, but also potentially finding gene mutations.

We were fortunate to adopt NanoFlares as the core detection technology because of its reproducibility, selectivity and sensitivity. The cumulative experience of the Mirkin laboratory in NanoFlare development contributed significantly to its reliable usage. Throughout the project, we maintained monthly remote meetings amongst the Mirkin, Paller and Xu laboratories. Receiving guidance as we transitioned the NanoFlares from detection in solution, to cells, to tissue and pre-clinical models was highly invaluable. Implementing the appropriate controls (e.g. non-coding sequences, uptake controls, housekeeping genes) and blinding procedures for animal studies was key to the smooth progression and instilling confidence in the project.

We envisioned that topically-applied NanoFlares would penetrate the skin barrier, enter the skin cells and interact with intracellular mRNA biomarkers (Figure 1a). We also demonstrated that NanoFlares successfully identified diseased cells (hypertrophic scar cells) from healthy cells and from diseased cells treated with transforming growth factor (TGF)-β inhibitors (Figure 1b).

Figure 1: Non-invasive disease diagnosis using skin-penetrating nanotechnology.  a) Skin-penetrating nanotechnology – NanoFlares interact with target biomarkers (mRNA) and become active, generating strong fluorescence signal. Upon topical application, NanoFlares penetrate the skin barrier and interacts with disease cellular biomarkers found within skin tissue. The fluorescence signal is then externally-collected. b) Imaging skin surfaces containing ‘Normal’, ‘Diseased’ and ‘diseased + treatment’ (treated with TGFβ inhibitors) cells. NanoFlares are topically applied on skin surface to detect disease biomarkers. Control biomarkers indicate the approximate number of cells (adapted from Figure 6b of our paper).

Bottlenecks and challenges. A common challenge in any study with translational potential is finding a suitable animal model. In fact, few animal models are suitable for studying abnormal scars. Because commonly used experimental small animals, such as mice and rats, heal differently from humans (by contraction rather than by re-epithelialization) and do not generate abnormal scarring to the extent of humans, finding suitable animal models for assessing NanoFlares was a challenge.

We began by injecting human scar fibroblasts into immunodeficient mice, but recognized that the difference in epidermal thickness between mouse skin and human (approximately 3 vs. 10 epidermal layers) was not a valid test of the ability of Nanoflares to penetrate the epidermis and reach dermal targets, as required for scar detection. Hence, viable human skin obtained from abdominoplasties was modified by introduction of normal, abnormal scar-derived, and disease-conditioned fibroblasts to create various tissue-based abnormal scar models. These had the advantages of overlying human epidermis and dermis, and a uniform number of loaded fibroblasts, allowing us to control the experimental conditions yet provide the appropriate challenge for testing NanoFlare transdermal delivery.

Prospects for transdermal nanotechnology diagnosis. Diagnosing skin disease and following disease progression without skin biopsy has huge potential benefits for patient, especially paediatric patients in whom biopsy is particularly problematic. Enabling diagnosis to be made through a procedure which, in part, could be self-administered, may increase diagnosis accessibility, speed diagnosis, allow serial tracking of response, and reduce cost. A future consideration is the logistics of signal detection noninvasively. To this end, we are currently working on detecting the NanoFlare technology with the use of handheld fluorescence detectors (Figure 2). By coupling these detectors with internet-enabled readout, diagnoses could even be made remotely, further facilitating patient care.

Most skin disease is not currently diagnosed or managed through gene expression assessment, and the use of NanoFlare technology introduces an entirely new paradigm for skin diagnosis and management. Our study represents just the first step for NanoFlare technology as a new tool for skin disease with great promise and much work to be done. 

Figure 2: Hand-held fluorescence detection on a clinical patient. a) Hand-held fluorescence detector applied to a suspected lesion on the patient’s skin (mole). b) Near-view of mole using the white-light channel. Scale bar represents: 2 mm. c) Fluorescence channel of bare skin (without probe). Specks of autofluorescence (AF) can be seen at the location of hair follicles. d) Probe fluorescence (PF) can be detected following topical application. A model probe (DiI fluorescence dye) was emulsified with Aquaphor® cream and smeared in the vicinity of the lesion.

This post was written by David Yeo, Amy S. Paller, Chad A. Mirkin and Chenjie Xu.

Our paper: Yeo, D. C. et al. Abnormal scar identification with spherical-nucleic-acid technology. Nat. Biomed. Eng. 2, 227–238 (2018) doi:10.1038/s41551-018-0218-x.

1. Hay, R. J. et al. The global burden of skin disease in 2010: an analysis of the prevalence and impact of skin conditions. J. Invest. Dermatol. 134, 1527-1534 (2014).
2. Bijlard, E. et al. Burden of Keloid Disease: A Cross-sectional Health-related Quality of Life Assessment. Acta Derm. Venereol. 97, 225-229 (2017).
3. Gauglitz, G. G., Korting, H. C., Pavicic, T., Ruzicka, T. & Jeschke, M. G. Hypertrophic Scarring and Keloids: Pathomechanisms and Current and Emerging Treatment Strategies. Mol. Med. 17, 113-125 (2011).

David Yeo

Research Fellow, Nanyang Technological University